BACKGROUND:
Joint arthrodesis often employs autograft to promote union; graft harvesting can lead to perioperative morbidity. A Canadian randomized controlled trial (RCT) demonstrated that recombinant human platelet-derived growth factor BB homodimer (rhPDGF-BB) combined with beta-tricalcium phosphate (β-TCP)-collagen was a safe, effective alternative to autograft. This multicenter North American RCT compared the safety and efficacy of rhPDGF-BB/β-TCP-collagen with autograft for ankle and hindfoot fusion. Subclassification using propensity scores (PS) incorporated patients from previous trials for enhanced statistical power for noninferiority testing and broader review of treatments.

METHODS:
Patients requiring ankle or hindfoot arthrodesis and supplemental bone graft were treated with rhPDGF-BB/β-TCP-collagen (n = 69) or autograft (n = 35). Outcomes included joint fusion on computed tomography (24 weeks), clinical healing status, visual analog scale (VAS) pain, Short-Form 12 (SF-12), American Orthopaedic Foot & Ankle Society (AOFAS) Ankle-Hindfoot Scale, and Foot Function Index (FFI) scores over 52 weeks. PS methodology addressed potential selection bias arising from pooling data among these patients and 2 previous RCTs with similar inclusion criteria, surgical techniques, graft harvest techniques, and outcomes. All 132 rhPDGF-BB/β-TCP-collagen-treated patients and 167 of 189 candidate autograft-treated controls were selected for comparison by an independent statistician blinded to outcomes.

RESULTS:
In the PS subclassification, 68.1% treatment patients and 68.4% controls achieved >50% osseous bridging at fusion sites. Clinical healing status was achieved in 84.8% of treated patients and 90.7% of controls at 52 weeks. Clinical, functional, and quality of life results demonstrated noninferiority of rhPDGF-BB/β-TCP-collagen to autograft. Safety-related outcomes were equivalent.

CONCLUSION:
PS subclassification analysis of 3 RCTs demonstrated that rhPDGF-BB/β-TCP-collagen was as effective as autograft for ankle and hindfoot fusions, with less pain and morbidity than treatment with autograft.

LEVEL OF EVIDENCE:
Level I, prospective randomized study.